k8s-Guideline

Welcome Kubernetes Learning Group

You can use the editor on GitHub to maintain and preview the content in Markdown files.

1. Setup a single node cluster at local

There are two solutions to setup a single node cluster on Windows 10. Docker for windows(Edge) is the easiest and recommended way to get Kubernetes at local. You have to make sure your pc’s internet connection is setup correctely (has access on www.google.com is a must).

• Setup with Docker for windows edge
• Setup with Minikube

1.1 Post installation check

To check whether k8s is created sucessfully on you local, You could run command kubectl config get-contexts

Below is going to show on the terminal if you setup k8s with Minikube.

Below is going to show on the terminal if you setup k8s with Docker for windows. Context might be missing, to fix it you have to add C:\Users\[replace with username]\.kube\config to HOME environment variable.

1.2 Enable Dashboard

By default, you won’t get the Kubernetes Dashboard. There are two ways to install Dashboard service:

Run Command: kubectl apply -f https://raw.githubusercontent.com/kubernetes/dashboard/master/aio/deploy/recommended/kubernetes-dashboard.yaml

Or Run Command: kubectl apply -f https://raw.githubusercontent.com/kubernetes/dashboard/master/src/deploy/alternative/kubernetes-dashboard.yaml

The first approach install a dashboard service which run in secure way - requst user login. The second approach install dashboard service which is totally insecure.

Run command: kubectl proxy

Output should be as below:

If you setup dashboard service as the frist approach. You can open link https://raw.githubusercontent.com/kubernetes/dashboard/master/src/deploy/alternative/kubernetes-dashboard.yaml.

Otherwise openlink: http://localhost:8001/api/v1/namespaces/kube-system/services/http:kubernetes-dashboard:/proxy/#!/cluster?namespaces=_all&namespace=default

If you setup dashboard service as the frist approach. you may need to login as below:

To get a token:

Run Command: kubectl -n kube-system get secret

Run Command: kubectl -n kube-system describe secret deployment-controller-token-wvvsz

2. Prepare a docker image

There is a simple node app docker file. build and push it to docker repository as below. Please replace ‘yxzhk’ as your own signature registered on docker hub. source code and configuration file is located here

docker build -t yxzhk/nodeapp .

docker push yxzhk/nodeapp

3. K8s deployment

Run command: kubectl apply -f kubia-deploy.yaml

Run: kubectl get pods

Run: kubectl get deployments

Check dashboard:

once K8s finish the deployment, check status again.

4. K8s service

4.1 Create Service

Run command: kubectl apply -f kubia-svc.yaml

Check results:

4.2 Access Service

Access service from a pod as bleow command format: Kubectl exec {pod name} – curl -s http://{service ip}

In our case, Run: Kubectl exec kubia-deployment-8497d7fdd9-hrccm -- curl -s http://10.105.16.224

Discovering service through environment variables: kubectl exec {pod name} env

In our case, Run: Kubectl exec kubia-deployment-8497d7fdd9-hrccm env

4.3 Access Service through FDQN

Connecting to the service through its FDQN(fully qualified domain name) servicename.default.svc.cluster.local

Run Command: kubectl exec -it kubia-deployment-8497d7fdd9-hrccm bash

Them Run: curl http://kubia.default.svc.cluster.local or

curl http://kubia.default or

curl http://kubia

4.4 Service Endpoints

5. K8s network

Source code can be found here

Run Command: kubectl apply -f deployment.yaml

deployment “service-test” created

run: kubectl get pods

service-test-cf849786d-gm9zl 1/1 Running 0 1h
service-test-cf849786d-k2bt7 1/1 Running 0 1h

run: kubectl get pods --selector=app=service_test_pod -o jsonpath='{.items[*].status.podIP}'

10.1.0.39 10.1.0.37

update podclient.yaml with one of above IP address.

run: kubectl apply -f podclient.yaml

pod “service-test-client1” created

run kubectl logs service-test-client1

HTTP/1.0 200 OK
Server: SimpleHTTP/0.6 Python/2.7.15
Date: Sun, 02 Dec 2018 01:39:12 GMT
Content-type: text/html
Content-Length: 47
Last-Modified: Sun, 02 Dec 2018 01:11:11 GMT
<p>Hello from service-test-cf849786d-k2bt7</p>

run kubectl apply -f service.yaml

service “service-test” created

run kubectl get service service-test

NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
service-test ClusterIP 10.107.143.159 <none> 80/TCP 1m

run kubectl apply -f serviceclient.yaml

pod “service-test-client2” created

run kubectl logs service-test-client2

HTTP/1.0 200 OK
Server: SimpleHTTP/0.6 Python/2.7.15
Date: Sun, 02 Dec 2018 01:46:41 GMT
Content-type: text/html
Content-Length: 47
Last-Modified: Sun, 02 Dec 2018 01:11:11 GMT
<p>Hello from service-test-cf849786d-k2bt7</p>

run kubectl describe services service-test

Name: service-test
Namespace: default
Labels: <none>
Annotations: kubectl.kubernetes.io/last-applied-configuration={“apiVersion”:”v1”,”kind”:”Service”,”metadata”:{“ann otations”:{},”name”:”service-test”,”namespace”:”default”},”spec”:{“ports”:[{“port”:80,”targetPort”:…
Selector: app=service_test_pod
Type: ClusterIP
IP: 10.107.143.159
Port: <unset> 80/TCP
TargetPort: http/TCP
Endpoints: 10.1.0.37:8080,10.1.0.39:8080
Session Affinity: None
Events: <none>

run kubectl apply -f serviceNodePort.yaml

run kubectl apply -f serviceLoadbalancer.yaml

run kubectl get service